A Deep Learning Based Real-time Pedestrian Recognition System
Author(s)
Sun, Tao
DownloadThesis PDF (7.089Mb)
Advisor
Stonebraker, Michael
Terms of use
Metadata
Show full item recordAbstract
This research project aims to develop an automated recognition system to understand the behavior of pedestrians in public videos. Such behavior prediction is helpful in a wide variety of security applications, such as loss prevention, missing people identification, etc. A deep learning-based detection framework and customized feature identification algorithms are combined to address this technical challenge. Our system is scalable to a large number of video feeds.
Date issued
2021-06Department
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science; System Design and Management Program.Publisher
Massachusetts Institute of Technology